Out of all the models of client/server computing, the HTTP-based model is the most popular one, where a client web browser talks to a CGI (or similar) script on the server. Virtually, any standalone software can be used in “distributive” manner by writing an appropriate CGI script for it that makes this model a model of choice in bioinformatics. But, this model has a serious flaw of lacking a client-side customization facility that results in redundant network activity and waste of user time.

SeWeR (Sequence analysis using Web Resources) is a DHTML interface that meets this need. It is integrated, highly customizable and can be used directly from the desktop or even can be hosted from a website. It has several unique features including a facility of automatic incorporation of a new service.

To meet the lack of a facility to generate graphics over the web, I wrote “Pastel” a graphics library in Perl. The API closely follows Java2D library with state of the art computational geometry and support for animation. It generates Scalable Vector Graphics, a revolutionary XML based vector format by W3C. BioSVG is a set of graphics modules in Perl written on top of Pastel with more data abstraction and higher level API to directly represent biological object. The work on Pastel and BioSVG is continuing. As a “proof of concept” I wrote a small application called “Savvy” using BioSVG and Pastel that draws plasmid maps. SeWeR with Savvy makes a complete sequence manipulation and analysis suit.